This project addresses the need for sensitive and predictive measures that can identify elderly persons at risk for falling, thereby preventing some of the human and societal costs of devastating injuries. Dynamic balance assessment methods will be combined with virtual-reality techniques to produce an innovative technology for evaluating balance in simulated real-world environments. Present techniques reproduce only simple environment conditions, a far cry from the challenging situations encountered in daily life. In Phase I, three-dimensional simulations of environmental situations in which the elderly experience instability will be developed. Postural responses and perceptions of young and elderly normal subjects will be tested. Feasibility of this approach will be evaluated by comparing responses of healthy elderly and elderly with a propensity to fall. In Phase II, interactive virtual-reality 'environments' with appropriate quantitative balance assessments will be created. A long-term goal is to fully realize the potential of virtual- reality technology, be developing functionally relevant balance testing techniques soundly based on psychophysics and clinical electro- physiological methods. By making the system user-friendly and cataloging quantitative information along with other health status measures, emerging abnormalities could be identified, shifting the emphasis from diagnosis to prevention.